EU AI Act compliance is not a checklist you bolt on after launch. It is an architecture decision. We build AI on EU-hosted infrastructure, with grounded RAG instead of free-text generation, mandatory human review for outbound communication, and audit trails as a default – the conditions the AI Act and GDPR ask for. Most of our production work is already designed this way.
Most of our production work is already AI-Act-aligned because the constraints overlap with our default architecture. GDV runs on Azure AI Search + GPT-4o via Microsoft AI Foundry in the EU, with grounded RAG over policy documents and audit logging built in. Evangelische Bank runs on bank-compliant Kubernetes, with role-based access and full audit trails. A leading donation platform has mandatory human review built into the email-agent workflow before any reply goes out.
For new projects, we run a compliance scan during the use case workshop: data classification, AI Act risk tier, GDPR posture, audit and human-in-the-loop requirements. The result feeds directly into the architecture, so nothing needs to be retrofitted later.
Grounded in your data
EU-hosted by default
Human-in-the-loop
Time to first project
first working version in four short sprints
Compliant by design
EU AI Act and GDPR posture built into the architecture, not retrofitted
Operable by your team
low-code architecture documented for non-technical operators
1. Use case and architecture
2. Build the working system
3. Hand over
We bring a decade of impact-tech experience and over 160 AI projects since 2019. Through our free AI for Impact course, more than 100,000 people have learned to use AI for the common good. We do not run inspiration days. We run scoping sessions and build engagements that ship – the way we have delivered AI for the organisations below:
A leading member network – production RAG chatbot serving 1,000+ HumHub members on n8n + Qdrant + GPT-4 via Microsoft EU, delivered in four sprints
GDV (German Insurers Association) – AI Knowledge Assistant over tens of thousands of policy documents for 400+ member companies
A leading German association – AI Member Platform combining chat-based discovery with traditional category filters
A leading donation platform – AI email agent with mandatory human review in pilot, on N8N and Azure OpenAI
Evangelische Bank – production sustainability scoring B2B tool aligning customers with the Paris 1.5°C climate target via XDC metrics
Default stack: n8n in Berlin, Qdrant in the EU, Azure OpenAI via Microsoft EU Sovereignty.
AI does not replace your operators or experts. It removes the lookup and drafting work. The judgement call stays with a human.
Mandatory human review for outbound communication. Anything that goes to a customer, member or beneficiary should be drafted by AI and signed off by a human.
Hallucination is the failure mode to design against. Grounded RAG with citations, not free-text generation. If a system cannot cite its source, treat its answer as a guess.
How long does a first project take?
Is your stack EU-compliant?
Can our non-technical team operate the system?
Do you handle regulated data?
Tell us about your use case. We will reply with a proposed architecture and a date, usually within a working day.
Simon Stegemann
Co-Founder and CEO
